Evaluation of 5B9 as a Calibrator or Expression of Results in Absorbance Values for the Diagnosis of Hit With a PF4/Heparin Specific Elisa.

BACKGROUND

Immunoassays detecting anti-PF4/H antibodies must be sensitive to exclude heparin-induced thrombocytopenia (HIT), and optical density (OD) values are useful for confirming HIT, but no calibration is currently available.

OBJECTIVES

To study the impact of OD values on the performance of the Asserachrom HPIA IgG in a cohort of patients with suspected HIT, and the value of a calibration performed with 5B9, a HIT monoclonal antibody.

METHODS

The HPIA IgG was performed in 170 patients with a high or intermediate probability of HIT. Results were expressed in OD450 or '5B9 equivalent' units, using a calibration done with 5B9. HIT was confirmed when HPIA and SRA/PF4 tests were positive.

RESULTS

HIT was excluded in 97 cases because HPIA and SRA/PF4 were negative. The HPIA was positive in 73 cases and HIT confirmed in 43 cases (SRA/PF4+). Applying an OD threshold of 1.05, the NPV and PPV of the test were 98% and 83%, respectively. Calibration of HPIA with 5B9 did not improve its performance, since similar AUC values (ROC curves) were obtained whether results were expressed in OD values or in equivalent units of 5B9. Bayesian analysis showed that in patients with an intermediate pre-test probability of HIT, the post-test probability equalled 1% when OD was less than 1, and 100% when OD was over 2.

CONCLUSION

5B9 as a calibrator failed to improve the performance of HPIA, but this assay can reliably exclude (when negative) or confirm HIT (when OD > 2), without requiring a functional assay.